| Summary: | Graduate Developing consistent and repeatable broad-scale methods for biodiversity modelling is an important goal to address as habitat loss, fragmentation and environmental degradation threaten our ability to maintain ecosystem and species diversity levels. Geospatial reviews of biodiversity monitoring have identified ecological indicators for the indirect mapping of species richness and ecosystem components modelling the processes controlling species distribution gradients. The goal of our research is to advance broad-scale biomonitoring by demonstrating how landscape-scale environmental indices can be used to model regional ecosystem and species diversity of British Columbia (BC), Canada. We meet our ecosystem-modelling goal by selecting and developing suitable ecological indicators from Earth observation data and terrain indices to represent the structure, composition and function of the environment, displaying both static and dynamic landscape processes of BC’s ecosystems. We regionalize the selected indirect indicators of biodiversity using a two-step clustering algorithm. The results display 16 ecologically distinct terrestrial ecosystems, 10 of which characterize the northern Boreal, coastal and Southern Interior mountain regions, and six represent the coastal lowlands, interior, Georgia Depression, Boreal and Taiga Plains of British Columbia. Comparing our classification to BC Ministry of Forests biogeoclimatic zone mapping, we find spatial similarity in the coastal, Taiga and Boreal Plains. Overall, our classification distinguishes a greater diversity of ecosystems in the mountainous regions of the province and greater homogeneity in the Central Interior where our landscape characteristics represent current productivity conditions. Our approach to ecosystem modelling supports legacy mapping by providing ecological information in under-sampled regions of BC and offers a method for consistent repeat modelling of ecosystem diversity to identify landscape change. To meet our species-modelling goal we ...
|
|---|